36 comments

[ 4.1 ms ] story [ 59.4 ms ] thread
They have come a very long way since the late 1990s when I was working there as a sysadmin and the data center was a couple of racks plus a tape robot in a back room of the Presidio office with an alarmingly slanted floor. The tape robot vendor had to come out and recalibrate the tape drives more often than I might have wanted.
It's frustrating that there's no way for people to (selectively) mirror the Internet Archive. $25-30M per year is a lot for a non-profit, but it's nothing for government agencies, or private corporations building Gen AI models.

I suspect having a few different teams competing (for funding) to provide mirrors would rapidly reduce the hardware cost too.

The density + power dissipation numbers quoted are extremely poor compared to enterprise storage. Hardware costs for the enterprise systems are also well below AWS (even assuming a short 5 year depreciation cycle on the enterprise boxes). Neither this article nor the vendors publish enough pricing information to do a thorough total cost of ownership analysis, but I can imagine someone the size of IA would not be paying normal margins to their vendors.

Don’t put any stock into the numbers in the article. They are mostly made up out of thin air.
> $25-30M per year is a lot for a non-profit

$25 million a year is not remotely a lot for a non-profit doing any kind of work at scale. Wikimedia's budget is about seven times that. My local Goodwill chapter has an annual budget greater than that.

Facilitating mirroring seems like it would open up another can of liability worms for the IA, as well as, potentially, for those mirroring it. For example, they recently lost an appeal of a major lawsuit brought by book publishers. And then there's the Wayback Machine itself; who knows what they've hoovered up from the public internet over the years? Would you be comfortable mirroring that?
> $25-30M per year is a lot for a non-profit

First, whether IA or any other large non-profit/charity. When you are in the double-digit/triple-digit multi-million bracket, you are no longer a non-profit/charity. You are in effect a business with a non-profit status.

Whether IA or any other large entity, when you get to that size, you don't benefit from the "oh they are a poor non-profit" mindset IMHO.

To be able to spend $25-30M a year, you clearly have to have a solid revenue stream both immediate and in the pipeline, that's Finances 101. Therefore you are in a privileged and enviable position that small non-profits can only dream of.

Second, I would be curious to know how much of that is of their own doing.

By that I mean, its sure cute to be located in the former Christian Science church on Funston Avenue in San Francisco’s Richmond District.

But they could most likely save a lot of money if they were located in a carrier-neutral facility.

For example, instead of paying for expensive external fiber lines (no doubt multiple, due to redundancy), they would have large amounts of capacity available through simple cross-connects.

Similar on energy. Are they benefiting from the same economies of scale that a carrier-neutral facility does ?

I am not saying the way they are doing it is wrong. I'm just genuinely curious to know what premium they are paying for doing it like they are.

This seems like a lot of zesty made-up assumptions.

And a lot of non-profits would be very very surprised to hear that once you cross the threshold of $9,999,999 costs, you are a business.

(comment deleted)
i dunno why i keep imagining something like ipfs could help something like the internet archive...
>And the rising popularity of generative AI adds yet another unpredictable dimension to the future survival of the public domain archive.

I'd say the nonprofit has found itself a profitable reason for its existence

Is this some kind of copypasted AI output? There are unformatted footnote numbers at the end of many sentences.
Maybe, but I was trying to find the original source of this article and couldn’t, at least not cursorily.
I already stopped when I saw the AI-gen image
The table also seems like the kind of thing that Gemini seems to generate a lot. "Here's a table that communicates almost no information! One of the rows is constant for each item."
The IA needs perhaps not just more money, but also more talented people, IMO. I worry that it has stagnated, from a tech pov.
They can offer a perk that literally no other tech job can offer: Someday have a statue of your likeness preserved in ceramic: https://www.atlasobscura.com/places/internet-archive-headqua...

"Inside the church's main room, with its still-intact pews, there are more than 120 ceramic sculptures of the Internet Archive's current and former employees, created by artist Nuala Creed and inspired by the statues of the Xian warriors in China."

We've hired a few dozen people over the past couple of years. We think they're pretty talented.
I think this was writen wholly by deep research.

It just reads like a clunky low quality article

IA is hosting a couple more of Rick Prelinger’s shows this month. Looking forward to visiting
Does any one know how the size of this compares to archive.today?
How long will it take for them to send the PetaBox to space?
This article is way too LLMey for my taste.
I love to imagine this is all a cover and the Internet Archive is located in a remote cave in northern Sweden and consists of a series of endlessly self replicating flash drives powered by the sun.
this is every data hoarders dream setup haha
Hate to be the guy in the comments complaining about the css, but the sides of the text of this article are cut off. It looks like I'm zoomed in, and there's no way I can see the first few columns of the text without going to Reader view. I'm on a modern iPhone using safari, accessibility settings font larger than usual.
It's an AI-generated article. It's going to be pretty terrible.
Is it still year 2006 and websites haven’t figured out responsive design?
Thanks for this, I've always wondered how the Archive operates but always ended up not searching.
Wow that piece of real-estate has to cost a bundle.
I have always wondered how archives manage to capture screenshots of paywalled pages like the New York Times or the Wall Street Journal. Do they have agreements with publishers, do their crawlers have special privileges to bypass detection, or do they use technology so advanced that companies cannot detect them?
(comment deleted)